Purpose While research has consistently identified an association between cannabis use and psychosis, few studies have examined this relationship in a polydrug context (i.e. combining cannabis with other illicit substances). The paper aims to discuss this issue. Design/methodology/approach The present study sought to examine the association between recreational drug use (cannabis only vs polydrug) and psychotic disorders. Analysis was conducted on a large, representative survey of young Danish people aged 24 (n=4,718). Participants completed self-report measures of lifetime drug use and this information was linked to the Danish psychiatric registry system. Findings Multivariate binary logistic regression analysis was used to examine the association between drug use (no drug use, cannabis only, cannabis and other drug) and ICD-10 psychotic disorders, while controlling for gender and parental history of psychosis. Compared with no drug use, the use of cannabis only did not increase the risk of psychosis while the odds ratio for cannabis and other drug were statistically significant. Research limitations/implications Psychosis risk may be associated with the cumulative effect of polydrug use. Practical implications Cannabis use may be a proxy for other drug use in research studies. Originality/value This study is innovative as it uses linked self-report and administrative data for a large sample. Administrative data were used to as an objective mental health status indicator.